Abstract
The article examines changing employment relations in Norwegian warehouses, and conceptualises the increasing use of temporary agency workers as a redrawing of workplace geographies. The empirical basis for the analysis is four qualitative warehouse workplace studies, including focus group and interview data. The theoretical framework of the article combines an adapted version of the territory-place-scale-network (TPSN) framework developed by Bob Jessop, Neil Brenner and Martin Jones with the concepts of labour control and labour agency. The analysis shows how a networked recruitment system based on Swedish labour migrants, mediated via temporary work agencies, encourage workers to work their way through levels of employment insecurity in order to secure permanent employment. The article argues that the blurring and redrawing of legal boundaries through labour hire can be understood as a territorial strategy of control that affects the workplace as a scale of justice for trade unions. Moreover, the analysis shows how managerial control is conditioned by workers’ individual, habitual and collective agency.
Introduction
Reading human geography literature one often comes across the workplace used as a subject of study, but far more unusual are attempts to reflect on the workplace as a geographical phenomenon. ‘The workplace at the millennium’, a special issue of Environment and Planning A in 2000, was commendable in raising the topic, and while we concur with the authors (Herod, 2000; Wills et al., 2000) that the workplace scale does connect with economic transformations and regulatory regimes far beyond the workplace, we argue that it is possible to reap different analytical rewards by taking the socio-material space where people work as a point of departure – rather than, say, the firm or the labour market. Changing workplace practices and innovative employment arrangements such as labour hire through temporary work agencies, team-based performance management and the automation of work tasks should not be understood solely as the effects of wider social currents, but also as drivers of the very same currents.
Labour geography’s development from the 1990s until today has gone hand in hand with the conceptualisation of scale in human geography. Through theoretically informed case studies, scale emerged as the central concept explaining how political actors sought to employ scalar strategies to alter vertical arrangements in the politics of work. Hence, the concept was used to understand the contradictions of labour internationalism (Castree, 2000; Herod, 1998) and the significance of the urban scale for living wage campaigns and other new forms of unionism (Savage, 2006; Walsh, 2000). While offering important insights, labour geographers were arguably culpable of a conceptual one-sidedness later critiqued by Jessop et al. (2008: 391) by favouring scale as ‘the primary basis around which other dimensions of socio-spatial relations are organized’. Taking this critique seriously, we seek to actively embrace a ‘polymorphy of sociospatial relations’ in our discussion of workplace change (Jessop et al., 2008: 392). In this article, we analyse how workers’ agency interacts with management practices of control at the scale of the workplace, by bringing them together in a refined version of the territory-place-scale-network (TPSN) framework suggested by Jessop et al. (2008).
From an empirical standpoint, the article investigates changing employment relations in Norwegian warehouses. Warehouses are focal nodes in global production and distribution networks where the demand for flexible staffing is particularly pronounced. In warehouse workplaces in the Oslo region, temporary work agencies operate extensively and increasingly, which is made possible by the national deregulation of labour hire and the implementation of EU law, and fuelled by short-term migration from Sweden. While the Norwegian trade union movement has attempted to resist and engage with legislative change at a national level, workers’ approaches to temporary agency staff in the workplace remain ambivalent.
The article starts with a discussion of how labour process theory and labour geography have understood expressions of control and agency in the workplace, before showing how a polymorphic socio-spatial framework can nuance these concepts. The latter part of the article illustrates the argument using material from focus group discussions and interviews in four different warehouse workplaces. Labour hire through temporary work agencies is explored through different conceptual angles: place embeddedness and the construction of workplace attachment; the networked recruitment of labour migrants; the territorial construction of legal boundaries in the workplace and, finally, the workplace as a scale of justice for trade unions. The article ends with a short conclusion trying to bring these dimensions together and demonstrate the implications for trade union politics.
Zooming in and branching out of the workplace
Our analysis of the workplace combines three strands of scholarship: the labour geography tradition and its exploration of the agency of workers (Coe and Jordhus-Lier, 2011; Hastings and MacKinnon, 2016; Herod, 2001); labour process theory and its understanding of managerial and collegial control (Edwards, 2010; Thompson and Smith, 2010) and, finally, Jessop et al.’s (2008) framework for theorising socio-spatial relations. Labour geography has until recently been mainly focused on various forms of labour struggles and campaigns (Castree, 2008; Herod, 2001; Wills, 2005) in an effort to demonstrate that workers are able to shape the economy through collective action. This focus has often been at the expense of individual and loosely organized expressions of agency (notable exceptions include Hastings and MacKinnon, 2016; Kiil and Knutsen, 2016; Rogaly, 2009). The present paper adds to current debates by attempting to re-embed collective and individual expressions of agency in the workplace. We argue that studying the micro-geographies of work brings out new dimensions of labour agency hitherto under-emphasised in geographical literature.
Agency is commonly understood as intentional, purposive acts to reach predefined goals or objectives. This is the instrumental and strategic side of agency. Yet, there is also another side to agency, referred to as ‘the habitual dimension of action’ (Emirbayer and Mische, 1998: 975): a less-reflexive, more-automated side of agency, or rather of praxis, of doing. Together with the collective/individual dimension, the strategic/habitual dimension forms a matrix within which different expressions of workplace agency can be located and understood. While labour geography has emphasised collective strategic action, such as trade unionism, the decision to quit the warehouse job to search for employment elsewhere can be characterized as both individual and strategic, while the competition between shifts responding to the same performance-based bonus system can be seen as collective agency grounded in habitual practices. We do not see these dimensions as dichotomous and mutually exclusive, but rather argue that they offer a better understanding for how agency is practiced in the workplace, and for how it interacts with various mechanisms of labour control.
This brings us the notion of control in the workplace, where we believe the labour process is a useful point of entry. A recent revival in labour process theory has been particularly interesting given its explicit embrace of geographical language (Edwards, 2010; Smith, 2010; Thompson and Smith, 2010). In particular, their exchange with geographers (Ellem, 2015; Rainnie et al., 2010) have helped labour process theorists better account for how spaces of reproduction and consumption beyond the workplace (such as living arrangements and transport) impact on the labour process (Rainnie et al., 2010). Moreover, attempts have been made to link the study of labour processes with the global production network approach (Newsome et al., 2015; Carswell and de Neve, 2013). Each of these can be seen as efforts to ‘branch out’ of the workplace. In this article, we suggest bringing concepts from labour process theory in contact with labour geography to achieve the converse – namely to ‘zoom in’ on the workplace. While labour geographers long have shown their ability to connect groups of workers with processes at multiple scales and across complex networks, their understanding of the social relations at work has often lacked the precision that characterises labour process theory.
The compatibility of labour process theory and labour geography has already been pointed out by others (Rainnie et al., 2010). Not only do they represent complementary analytical foci – of zooming in and branching out, respectively – but they share a neo-Marxist theoretical foundation. Ever since the seminal work of Braverman (1974), industrial sociologists have sought to explore the dialectic between managerial control and labor resistance at the workplace level. Rather than assuming open conflict, this approach views the social coordination of consent and resistance as a structured antagonism produced through daily working routines and habitual interaction (Burawoy, 1979; Edwards, 2010). The conversion of workers into labour power is at the source of this antagonism. Labour power is realised in the workplace, subject to the negotiation of work effort and highly dependent on the contingencies of different managerial control systems (Thompson and Smith, 2010).
Managerial control shapes, and is in turn shaped by, expressions of labour agency. Historically, the co-location of workers has provided the social capital and geographical proximity required for collective action and identity formation (cf. Nicholls, 2009). However, modern workplaces are more heterogeneous, segmented, fluid and complex than an ideal-type factory. Collective identity formation will therefore need to bridge different social and cultural identities, but also an increasing number of employee categories and contractual relationships. Moreover, while labour process theory tends to take the employment relationship as an analytical starting point, this article focuses on workplaces dominated by labour hire. The warehouse workforce is co-located in the contained and controlled production space of a warehouse, but employment relationships are being fragmented through labour market mediation. Here labour geography’s ability to zoom out is useful in order to see the labour process in a wider context of mobility practices and the rescaling of legislation.
The polymorphy of the workplace
Jessop et al. (2008) advocate the use of multiple socio-spatial concepts for analysing social phenomena geographically. To overcome geography’s tendency of one-dimensionalism throughout its conceptual development, they suggest a division of labour between four key socio-spatial concepts – territory (T), place (P), scale (S) and network (N) – also known as the TPSN framework. Intended as an approach for analysing macro-processes of restructuring in the political economy and, according to the authors, particularly useful for a studies of change and controversy, we would argue that the framework also has merit for understanding the micro-geographies of work and the contested practice of labour hire. To give a better explanation of how the concepts scale, network, territory and place apply to the workplace, each of these will be given a brief consideration in the following section.
The workplace as scale
Jessop et al. (2008) argue that the scale concept holds most precision when used relationally to describe vertical differentiation in the political economy. Hence, they advocate for a more narrow use of the term than some geographers, who have used scale as a lens through which to explain ontological questions about the social ordering of production and collective consumption in capitalist societies (e.g. Marston, 2000). Within labour geography, strong interest in urban labour markets and global production networks has sometimes trained the attention away from the workplace as a scalar configuration, but the workplace has not been overlooked. Gough’s (2003) ‘internal relations’ approach, for example, offers a geographer’s take on labour process theory by conceiving the workplace as an expression of the capital–labour relation: This relation underpins global circulation of capital as well as local cultures of industrial relations, and the interaction between workers and management in capitalist firms. Gough (2003) argues that the agency of workers in the workplace is instrumental in shaping control, consent and resistance. Importantly, workplace dynamics are also shaped by non-capitalist relations between workers, and between workers, management and other social actors.
Others have focused on the scalar arrangements of labour legislation. Weller (2007) is concerned with how workplaces are relationally co-constituted with scales of jurisdiction within national labour markets and supra-national regulatory scales with an impact on labour legislation. This is also evident in Wills’ (2001) study of the European Works Councils system. Forms of regulation serve as meso-level political scales mediating the globalisation strategies of expansive capitalist firms and the local interest of workers (Herod and Wright, 2008). As we will see, this is also the case in how EU directives have affected workplace dynamics in Norwegian warehouses.
The workplace accommodates relations between multiple scales: from corporate strategies and state policy-making to the household, and to the scale of the body (Vanoutrive, 2012). McDowell’s (2009) work has been important in demonstrating how labour is embodied in the workplace: while some workers are expected to be seen, and judged by their appearance, other workers are rendered invisible, and while workers’ bodies in the service sector are supposed to be good-looking and desirable, the bodies of others are incorporated into the production process as hard muscle or instruments of nimble precision.
Following from this, the scale concept helps us not only to see the workplace as part of hierarchical relations between individual workers’ agency and the structures of capital and state regulation – but to appreciate how scalar reconfiguration at other scales (such as the national deregulation of labour hire) impacts on workplaces. But these changes are implicated in yet other changes affecting workplaces that are not as easily captured by the language of scale. One of them is the recasting of employer responsibility resulting from the use of temporary work agencies, creating new and invisible boundaries in the workplace. Here, we build on Jessop et al. (2008) to argue that socio-spatial relations exhibit other qualities than those who can be reduced to hierarchies and vertical difference. The authors propose that we can get a better understanding of each of the socio-spatial dimensions by treating them as a field structured by other geographical principles. To explore the scale concept in this manner, we will investigate how the workplace scale is co-constituted by enclosure and territorial control (territory), by interconnectivity (network) and by geographical proximity and social embeddedness (place). Let us visit these in turn.
Territoriality at work
Jessop et al. (2008) understand territory as an expression of enclosure and bordering. The territoriality of workplaces in part is expressed through external boundaries, be they of a material character, such as factory walls or property boundaries, legally expressed through the employment relationship, or as a social community at work. The boundaries of the workplace are ever changing, and observers argue that workplace territorialities are redrawn through technological change across the globe. For instance, more women enter formal workplaces where they can form supportive relations of friendship beyond the home (McDowell, 2009; McDowell et al., 2007). There is also a reverse process taking place, with subcontracted manufacturing being outsourced from factories to private homes (Carswell and de Neve, 2013). The commodification of care work also makes low-paid women perform domestic duties far away from their own homes (Pratt, 1999). For higher income groups, ‘new forms of technology enable workers to free themselves from the fixed boundaries of their workplace’ (McDowell, 2009: 15). While ‘the workplace’ has been fragmented and dispersed as a site for waged labour in complex ways, we would argue that this should not lead scholars to disregard how concrete workplaces still represent the tangible, social and institutional boundaries of work for millions of workers.
In this article, we are particularly interested in how workplace territoriality is expressed as a negotiation between the legal and social dimension within the material confines of the warehouse. Norwegian warehouses witness an ongoing externalisation of work, where tasks are performed by individuals without an employment relation with the client firm – a commonly used strategy by employers as a means to deal with fluctuating demand (Holst, 2014; Spermann, 2011). These practices create ambiguous positionalities in the workplaces, where individuals can find themselves within the boundaries of the material workplace, without being included in the legal and social community of work. These geographies are evident in the language often used to describe these individuals as either ‘outsourced’ or ‘in-house’ workers.
The social and legal boundaries of the workplace can be contested. Who is on the inside and who is on the outside is always a matter of debate and negotiation. The construction of insider and outsider identities are often co-constructed by many actors (employees, supervisors, managers, union representatives and labour market intermediaries) at once. While employer-initiated outsourcing of particular worker groups or departments in the workplace fragments employment relations (Jordhus-Lier, 2014; Reimer, 2003), fragmentation can be partially offset by efforts to maintain workplace cohesion by workers themselves, or by regulation which bars unequal treatment of workers in the workplace. Hence, perceived and experienced boundaries might not match legal and institutional boundaries.
While TPSN framework conceives of territoriality in a way that helps us investigate these contested boundaries, Jessop et al. (2008) tend to reduce the territorial dimension to the construction of external boundaries. An equally important expression of territoriality is the parallel acts of ‘delimiting and asserting control over a geographical area’ (Sack, 1983: 56, italics added). Hence, we also take territoriality to refer to practices using the materiality of landscapes or, in our case, the materiality of workplaces (see also Baldry, 1999), to exert indirect power over people. Crang’s (2004) vivid account of working in a busy restaurant is one of the more well-known illustrations of this in the human geography literature. He emphasises the performative aspects of service work, and portrays the workplace as a stage where service personnel negotiate expectations from customers and managers. The drama unfolding is structured by the division between kitchen and restaurant, between territories of different waiters and in the movements of service staff (see also McMorran, 2012). Managers enforce workplace territoriality through technologies of control, and the warehouse sector display a wide range of such technologies, including pick-by-voice systems, surveillance and digital systems for monitoring work effort.
The networked workplace
The network concept captures the interconnectivity and interdependence of socio-spatial relations (Jessop et al., 2008). These principles complement and add to the concepts of scale/hierarchy and territory/boundary by encouraging us to appreciate the fluid and dynamic character of workplaces. While the workplace is seldom explicitly discussed in a network perspective, recent contributions by McDowell (2009) and McDowell et al. (2007) provides us with a start. By explicitly drawing on multiple spatial concepts simultaneously, her analytical framework reminisces of, but does not completely overlap with, the TPSN framework. She does not see the workplace as ‘a single example at a particular moment’, but rather as ‘constituted through and by interconnections’ McDowell (2009: 221). Workplaces are therefore viewed as local specificities produced by the coming together of ‘transnational movements, supranational and national legislation, regional and local variations in labour supply and demand [and] the uneven distribution of different types of work’ (McDowell, 2009: 218).
In the social sciences, the interconnectivity of workplaces have been described in different ways: as nodes in global production networks (Coe, 2013); as organisations mediating the flows of capital, information and symbols (Castells, 1996) and as points of temporary fixity in the mobilities of consumers and labour (Urry, 2007; Vanoutrive, 2012). The warehouses in our study form nodes in vast retail networks through wholesale, supply and distribution, and cater to production and distribution networks across the world. Flows of goods intersect just-in-time at the warehouse workplace, where workers load and unload, pick and pack the goods, before they are sent off to other nodes in the network.
However, the warehouses also depend on complex networks of mobile labour in order to function. In particular, young and migrant workers from various places in Sweden arrive to work in this sector. In our analysis, we are particularly interested in how the recruitment of migrant labour through labour market intermediaries constitutes an important network dimension shaping the social relations in the workplace, and creating interdependencies between in-house and temporary staff, client firms and temporary work agencies.
Placing work
The previous three spatial concepts have directed our attention towards how workplaces are hierarchically constituted, how they are enclosed materially, legally and socially and how they are connected to other social networks. But neither scale, nor territory, nor network provides us with a language to appreciate how workplaces are social spaces where workers come together to work – and what that entails for the politics of labour. This brings us to the place dimension of the TPSN framework, in which Jessop et al. (2008) see embeddedness and proximity as key principles.
Workers are embedded in their workplaces in various ways. In labour process theory, the workplace is understood as the co-location of the labour process (Braverman, 1974), which naturally implies the co-location of workers. Co-location disciplines workers as well as capital, with both sides being at the mercy of the same social infrastructure and the same technological opportunities and constraints. Workplaces accommodate this fundamental contradiction of the economic system, which encourages management to pursue technological and organisational restructuring in place or the relocation of production to other places (Gough, 2003). When faced with the militancy of place-based worker cultures, management can also attempt to bring in alternate sources of labour power (Ellem, 2015). Managerial strategies of control and recruitment regimes are thus working in tandem with local workplace cultures of employment.
As an arena of proximate social relations – as ‘a place in which peers, superiors, subordinates, managers and bosses interact’ (McDowell, 2009: 7) – the place dimension is highly tangible, perceived and experienced by those present in very concrete ways. As a consequence, workplaces can generate a strong ‘sense of place’ (or to rephrase, a ‘sense of workplace’) which for some workers might be based on loyalty towards their co-workers, or an attachment to the building in which they work, or as a sense of pride about the institution or brand they represent. Because workers are differentially embedded in the social practices of the workplace, the micro-geographies of work can display core-periphery relations. These might resonate with workers’ position in the labour market (cf. Atkinson, 1984), but are also experienced through social identities in the workplace and expressed through the performative aspects of work (Byoung-Hoon and Frenkel, 2004; Koene and van Riemsdijk, 2005; McMorran, 2012). By way of summary, embeddedness and co-location in the workplace enable the building of solidarity through trade unionism. While labour geography has problematised the workplace as a scale of justice, vis-à-vis other political scales, it should also be understood as a space of justice – performed through habitual and collective action.
Socio-spatial expressions of workplace control and agency
By seeing all four dimensions in relation to each other, we argue that labour geography can further its conceptualisation of political scale and scalar strategies (Hastings and MacKinnon, 2016; Tufts, 2007; Weller, 2007). In particular, it invites us to see the workplace as a political scale built from the ground up through processes of legal demarcation and internal control (territory), through social embeddedness and differentiation (place), and through the many interconnections of its workforce (network). Based on the analysis of warehouse workplaces to be presented in this article, we demonstrate in Table 1 how expressions of control and agency in the workplace are articulated in related, yet analytically distinct, ways, and how these can be organised according to the principles of socio-spatial structuration as suggested by Jessop et al. (2008). While we pay particular attention to practices of labour hire and migration, because these emerge as significant in the cases studied, we believe a polymorphic approach also can be useful in examining contested restructuring in other kinds of workplaces.
A TPSN approach to control and agency in warehouse workplaces.
TPSN: territory-place-scale-network.
While the following sections will demonstrate how these dimensions intersect in more detail, we will take time here to reflect on what we believe are four overarching strengths with this model. First, a polymorphic view of the workplace allows us to better detect socio-spatial challenges to workers in contexts where it is not ‘scale per se that is the prime object of contestation’ (MacKinnon, 2011: 22). Employers’ use of temporary work agencies to limit their legal employment responsibilities serves as an example, which in the workplace is perceived as a territorial rather than scalar dispute. Second, the table encourages us to see geographies of control and geographies of worker agency in combination. Hence, management’s attempt to redraw workplace relations through externalising the employer role is conditioned by workers’ own actions, which might serve to challenge or reinforce a boundary between in-house and hired staff. Third, and related, the concept of labour agency is not confined to collective strategic action, but includes habitual practices and individual strategies. Thus, we can better appreciate how worker exert contradictory pressures on workplace geographies, even when the formal trade union policy is to construct broad-based solidarity at work. Finally, the different socio-spatial dimensions in the matrix influence each other in sophisticated ways. Again, when employers seek to distance themselves from direct employer responsibility in the workplace through labour hire, this legal boundary can be renegotiated through regulation that enforces equal treatment and holds client firms jointly liable to hired staff. Based on this analytical framework, we will now explore the workplace geographies in Norwegian warehouses, as employment practices in this sector are undergone changes.
Restructuring and re-regulating warehouse employment in the Oslo region
Warehouses around Oslo have undergone complex transformations in recent years. A long trend of automatisation and the optimalisation of just-in-time systems have increased firms’ imperative for flexible solutions to their manual labour demand (Baker and Halim, 2007). Warehouses need to cater for and adjust to seasonal and weekly variation through a buffer of labour, making these workplaces striking examples of Atkinson’s (1984) numerical flexibility in practice. Skill requirements usually are low and that there are few differences in work tasks between temporary or hired workers on the one hand and permanent staff on the other. Not surprisingly, in Norway, warehouse and logistics firms are second only to the construction industry as recipients of workers from temporary work agencies (TWAs) in 2016 (NHO Service, 2016). Available data for the period 2008 and 2017 indicates a general decrease in the number of direct employees in warehouse, logistics and transport services, and a concomitant increase of the share and absolute numbers of hired workers in the same industry. This picture is also corroborated by trade union representatives we interviewed.
Warehouse workplaces are regulated through multiple scales of legislation, including supranational legislative directives, national laws and collective agreements (Rutherford and Gertler, 2002). One of the most important areas of regulation is the use of hired staff (Vosko, 2009). In Norway, labour hire was prohibited by law until the year 2000, and the role of TWAs has continued to be a highly contested issue in Norwegian debate (Jordhus-Lier et al., 2015). For warehouses, both the national deregulation of labour hire in 2000 and the recent implementation of the EU Directive on Temporary Agency Work (2008/104/EC) have changed how client firms compose their workforce. On the one hand, it has legitimised and normalised the TWA industry, while on the other hand, enforced a principle of equal treatment between direct employees and temporary agency workers (TAWs) in the workplace.
Despite the legal conditions and restrictions that apply through Norwegian labour law and collective agreements, we argue that this relatively new practice of labour hire has been an important driver of changing employment relations in warehouse workplaces for two main reasons. The first reason is that it is easy to monitor productivity or effort (e.g. number of picks per day). Here, management uses the monitoring of TWA workers’ performance as a screening device for (temporary) contract extension or permanent recruitment. The performance of TAWs also function as a direct comparison against the performance of permanent staff, in this way exerting pressure on all workers. Secondly, the increased use of TAWs leads to a fragmentation of the worker collective, dividing into more distinct core and peripheries of the workplace (Kalleberg, 2003) where reciprocities between groups of workers become less clear.
Moreover, warehouse workplaces reflect how labour mobility changes employment relations, particularly through practices of labour hire. Approximately 40% of TAWs in Norway have a different country background, and in a recent survey, 28% of Swedish TAWs worked in the transport, logistics and warehouse sector (Ellingsen et al., 2018). The inflow of Swedish labour migrants thus represents a vital network dimension of temporary work in the warehouse industry. The number of Swedish workers in Norway rose from 25,571 in 2003 to 55,103 in 2013 (Statistics Norway, 2015). Swedes hold several linguistic and cultural advantages in the Norwegian labour market, and Norwegian employers prefer them to other migrants in frontline customer services and in performing team-based and relatively complex logistical tasks in warehouses. As the need for large industrial spaces tends to push warehouse complexes out of urban centres, the workforce typically consists of two groups: permanent in-house employees of mainly Norwegian background, often local to the area, and temporary workers who tend to be Swedish short-term or long-term migrants, many of which reside in central Oslo.
In order to explore the socio-spatial dimensions of workplaces in this sector, we employ a multi-faceted qualitative data collection strategy based on a selection of four workplaces (see Table 2). The selection comprises a variety of sectoral specialisations, employment practices and levels of unionisation. We interviewed managers, supervisors and representatives of trade unions and TWAs in each workplace, but the main data collection strategy was a series of stratified focus groups with internal (mainly Norwegian) and external (mainly Swedish) employees. All in all, nine focus group interviews were conducted and the number of interviewees amounted to 37 (out of which 24 were Swedes).
Workplace dynamics in four warehouses.
TAWs: temporary agency workers.
Focus groups can by no means substitute the wealth of data in comparative surveys, or the empirical richness of participant observation. Still, we concur with Goss and Leinbach (1996: 117, 118) in that the focus groups method is a relatively effective way to cover multiple cases while allowing participants to ‘describe their experiences in locally relevant terms’, allowing researchers to observe ‘the formation of a temporary social structure that is a microcosm of the larger context’. In each of the focus groups, we used a visualisation exercise to prompt participants to reflect on their attachment to the workplace community, and their perceived insider/outsider status. In the focus group discussion, this exercise both served as an icebreaker and as an effective visual aid to facilitate a conversation about how competition, flexibility and externalisation are experienced and performed on the warehouse floor. In the following sections, we will analyse some of the key findings from this study by using the framework above.
Workplace embeddedness
We begin this analysis by showing how warehouse workers were anchored in their places of work. In legal terms, workers were formally tied to the workplace through an employment contract with the client firm or a TWA, but their embeddedness is also mediated through social relations with management and fellow employees. In conversations with workers, the socially constructed and subjectively experienced character of this embeddedness emerged. When asked how different participants felt a sense of attachment to the workplace, both permanent employees and external agency workers stressed social reciprocity as essential for their sense of belonging, but the relative importance attributed to this factor differed according to job security and the length of their employment at a particular workplace. This was clearly illustrated during the focus group discussions, when we asked participants to place a marker on a figure representing the ‘workplace community’ (see Figure 1(a) to (d)).
Sense of workplace belonging among permanent workers (PWs) and temporary agency workers (TAWs) in Warehouse A (triangles), B (squares), C (circles) and D (crosses). (a) perception of present position (PW), (b) perception of present position (TAW), (c) perception of position in retrospect (PW) and (d) perception of position in employer firm (TAW).
Figure 1(a) and (b) shows how workers assessed their present position in the workplace community for permanent workers (a) and TAWs (b), respectively. Unsurprisingly, permanent workers tended to assess their position closer to the perceived core of the workplace. Legal status is thus important in determining how workers experience their embeddedness in place. Permanent workers cited diverse reasons for their core position, ranging from being given responsibility by management or in the local union, meeting colleagues outside work hours, or even pride in the company brand. But there were also external workers who placed themselves in the core, even after having been working in the warehouse for less than a month. For them, social inclusion among colleagues and amicable relations with their boss are put forward as key reasons for the position. WORKER: You do feel at home, I mean, I’m enjoying it here and everything, but as you say, everyone is replaceable. INTERVIEWER: It is interesting where you place yourself, I mean you haven’t worked here that long? WORKER: Yes, but […] it depends on how you understand the question. I mean, I’m having such a good time here, I should be in the middle. But at the same time I do not really feel at home anywhere. (Focus group with external agency workers, Warehouse A)
Workers actively embedded themselves in the workplace community by adapting to existing norms and habitual practices: ‘fitting in’, adopting the ‘team spirit’ and performing in accordance with the expectations mapped onto certain categories of workers, were important (Zampoukos et al., 2018). Responding to, and engaging in, peer pressure to attain bonuses for team performance produced a paradoxical space of camaraderie and competition in the warehouse. Core-periphery positions were also related to nationality. Being Norwegian often meant being part of the permanent core group of workers. However, several workplaces had undergone dramatic shifts in recent years as Swedish workers had become more numerous, and taking up more central positions in the workplace. Some workers also talked about ‘working on the Swedish shift’, in which case being Swedish directly facilitated social inclusion.
As place embeddedness evolves over time, the temporal dimension warrants further reflection. In the four warehouses we studied, the recruitment practice resembled a try-and-hire system where most workers had gone through a period as TAWs in the same workplace before being offered a permanent position. This allowed us to ask permanent workers about their experiences in retrospect (Figure 1(c)). Here, the discrepancy between position as permanent and agency worker was more articulated. Participants cited lack of formal job security as a key reason. I have to say I placed it all the way out there, not because of the treatment I got from the company, but purely because of the anxiety I had for suddenly standing there without a job and because my contact at [the TWA] was a jerk. (Previously external agency worker, Warehouse C) I mean, you are nobody in that firm. Here you are in a workplace, you meet people every day. To create a relationship with [the TWA] is like, timesheet and salary. (External agency worker, Warehouse A)
Networked recruitment of Swedish migrants
Workplace embeddedness was partially conditioned by the insider/outsider status produced through labour hire practices, but was also shaped by the networked character of recruitment in these warehouses. In other words, the manner in which workers arrived at the workplace influenced management’s ability to control their integration in the labour process – as well as workers’ ability and willingness to resist this control. Different groups of migrant labour make up an important part of the workforce in Norwegian logistics firms. While Eastern Europeans often worked as truck drivers, the main migrant group doing the labour-intensive pick and pack work within the warehouses were Swedes. Migrant recruitment networks ran through all the workplaces studied: through personal networks, sometimes facilitated by Swedish government agencies, and through the transnational recruitment strategies of temporary work agencies.
The flow of short-term labour from Sweden to Norway was to a limited extent facilitated by temporary work agencies. One Swedish worker explained that a TWA was the channel through which he got a temporary job in Oslo in 2010, when labour demand in the warehouse sector was high. The worker explained that he took the role of a ‘scout’ once established in Norway: I had the interview with [the TWA] in Gothenburg on Wednesday, and on Friday they called to ask if I could start work in Oslo on Monday [the following week] and that’s what I did. […] after that, I became a leader of sorts for other Swedes coming here in the same way. (Former external worker, Workplace A) If a person from Sweden gets a job with us, numerous applications from people with similar profiles, and from the same geographical area come our way […] they must be talking through their networks […] this is especially so if they think they’re being taken good care of. (TWA manager, Oslo) I guess that […] when you’ve travelled to Oslo, or moved away from your home […] just that step […] to me it was a big thing just going. (Swedish external worker, Warehouse D) I think our experiences say that those who need a lot of help getting to Norway, do not function as well in the labour market […] don’t make it […] If you need help with moving, then you’re not ‘job-ready’. (TWA manager, Oslo)
To draw on Smith’s (2006; 2010) terminology, this networked recruitment served to evaluate workers’ mobility power (i.e. ability to move) as well as their effort power (i.e. their ability to work). Effort power of the warehouse workforce was maximised through bonus systems and technologies of control, but also through associating Swedish TAWs with a stereotype of being hard-working and effective. These assumptions were not only expressed by managers. Our focus group discussions indicated that these expectations also were internalised by the Swedish workers (Zampoukos et al., 2018; see also McDowell et al., 2007). We are known to work a lot of overtime, and to work hard. That’s what they expect of us when they hire us, that we are going to give that little extra. That’s my impression, at least. […] We are not here to live, we are here to work. That’s why we travel to Norway. (Swedish external worker, Warehouse B)
Territorial expressions of labour hire
It could be argued that the way workers were recruited to, and embedded in, the workplace constructed a community of interest between different groups of employees and employers. But the attempts by the latter to limit the extent of their employer responsibility through labour hire created tensions. We found that the increasing use of TWAs in warehouses produced new boundaries in the workplace, and that these divisions can be understood through the language of territory. In legal terms, a worker hired on a temporary basis through a commercial agreement between the client firm and her/his employer can be understood as an outsider in the workplace, while those with a direct employment relation to the place of work are insiders. Across the four warehouses studied, however, this legal boundary was practiced and experienced in different ways.
Two dimensions in particular appear to influence this process: first, the different flexibility strategies pursued by firm management, and second, the relative power position of the labour market intermediary vis-à-vis the client firm. Managers presented their main labour hire strategy in distinct ways: to meet seasonal or short-term demand, as a more permanent buffer strategy, as a form of recruitment (see above) or as a strategy during transition periods of restructuring. These strategies impacted differently on how permanent workers and TAWs related to each other in the workplace. As can be read from Table 2, hiring external labour for recruitment purposes seemed to mask formal boundaries in the workplace. Buffer and restructuring strategies could also have this effect, when TAWs experienced that they did not serve a temporary demand, and were allowed developed stronger social ties to the permanent workers and harbour expectations of permanent positions. Short-term labour hire had the opposite effect, particularly when combined with clear and predictable practices of direct recruitment by the client firm.
Another factor was the relative power position of the TWA to the client firm (see Table 2). In Warehouse A, where one single TWA had developed a relationship of trust with the client firm, TAWs were regularly offered permanent positions. Here, the boundary between permanent and external workers became less articulated. An HR manager justified this as a means to increase performance, but also as part of Warehouse A’s recruitment strategy. Hence, they combined their need for a buffer with a screening process through which external agency workers were considered eligible for a permanent contract: I expect the agency worker to give more […] Entering the labour market, they have to fight for a permanent job. I expect them to contribute that little bit extra […] This is how we recruit, really. (Mid-level manager, Warehouse A) I don’t really feel insecure about getting a new contract […] as long as I do a good job. (External agency worker, Warehouse A) Having multiple TWAs enhances our chances of getting enough good workers […] then, two agencies will fight it out between them to deliver the best people at any given time. If a group of [agency] workers are sub-standard, we won’t hire many from that particular agency the next time. (HR manager, Warehouse D) The TWAs say that if the agency workers are notified too early about whether the contract is extended or not, they lose motivation or don’t show up for work […] This is why we agree on a policy in which they [agency workers] are notified only a few days in advance. (HR manager, Warehouse D) It happened sort of automatically, or what can I say? But sure, once in a while you become worried. […] You don’t know if you have a job in 14 days. It is… modern slavery. You can cite me on that. (Agency worker, Warehouse D) If you tell 40 agency workers that they are considering 10 permanent positions […] of course it will be about who works the hardest. You put the carrot in the middle. (Former external agency worker, Workplace A)
Individual and team performance was constantly monitored. Shifts would be expected to reach a certain target, sometimes in competition for bonuses. Team leaders in particular received a bonus based on team performance. Informants in Warehouse C described this dual system of control and incentives as an important part of workplace relations. Workers came to pay more attention to the performance of their co-workers: [Local managers] are expected to show good numbers from the pickers in their teams to their superiors. Every manager has to show something to their boss, that we can deliver. Then it’s very bad to […] have a person that does not perform […] does not show good enough numbers. (Worker, Warehouse C)
Scales of justice
Thus far, the discussion about the agency of workers has been largely restricted to individual decisions (e.g. of moving) and the habitual practices (e.g. of working harder) that develop among colleagues. But labour hire has also affected the possibilities of collective agency in the workplace. Labour geographers have emphasised the importance of articulating solidarity at social scales that people experience as tangible and relevant in their lives, such as workplaces and neighbourhoods. When workplace organisation becomes the primary political scale, however, it can easily regress into ‘militant particularisms’, or appear as a local ‘scale of justice’ for capital–labour alliances (Gough, 2010).
The practice of labour hire has increased the significance of the workplace scale in transport and logistics. Local union representatives in the workplace are now authorised to enforced two key legal principles in practice: the principle of permanent employment as the main norm, as enshrined in Norwegian labour law, and the principle of equal treatment (including pay) for in-house and hired staff, as secured by the EU Directive on Temporary Agency Work. Management is legally obliged to consult the local union representatives on issues of labour hire. As regulatory implementation rely on the capacity of the local union organisation, union representatives need to be both knowledgeable about corporate practices as well as willing and able to confront management.
But as this analysis has shown, the workplace scale is conditioned by its territorial, networked and place-based characteristics. Local trade unions struggle to function as effective representatives unless they prove able to build solidarity across a fragmented workforce by strengthening a sense of workplace attachment and actively challenge the legal boundaries produced by labour hire. As such, the workplace is not only a scale of justice, but a ‘space of justice’ that local unions take part in constructing. A representative of the national trade union (Norsk transportarbeiderforbund) described the warehouse floor as an important recruitment space in the sector: while the truck driver workforce is mobile and dispersed, warehouse workers meet in large social spaces providing a basis for solidarity. Different unions have sought to establish strong local organisations in warehouses, depending on which sector the warehouse serves. In our interviews and focus groups, workers’ experiences with their local trade unions varied across the four warehouses (see Table 2).
In three of the workplace cases, the union representatives can best be characterised as co-managers of flexibilisation processes initiated by management. In Warehouse A, the above-mentioned long-term arrangement with a single temporary work agency was supported by the local union representative as well as by many permanent workers. One TAW expressed that the union representative ‘was on board with management’ in questions of labour hire. She did not feel represented by the union.
In Warehouses B and D, management defended their extensive use of labour hire by referring to the ongoing restructuring. At the time of research, union representatives appeared incapable of extending the number of direct and permanent positions. The union and the firm has met many times, and we say “you’ve got to cut the temps”, but the firm says «no, this is just temporary, in four months they are gone». […] And then the union gives them another four months, and they come back and say “hey, you see… now, we only need another three months”, you understand? And that’s how they buy themselves out. (Worker, Warehouse D) I am very opposed to temporary work agencies. They are a business based on breaking the law. There shouldn’t be any temporary positions in Norway, bar a few exceptions that you probably know. (Union representative, Warehouse C)
Conclusion
This article has investigated the workplace geographies of Norwegian warehouses. These workplaces have undergone considerable changes in their employment relations, not the least through the increased use of temporary work agencies. We have employed the TPSN framework, originally proposed by Jessop et al. (2008), to argue that labour geographers’ conceptualisation of scale can benefit from engaging with the socio-spatial concepts of territory, place and network. Our polymorphic approach favours an understanding of the workplace scale as built from the bottom up, emphasising the social relations between different groups of workers and management, without losing sight of the multi-scalar nature of labour legislation and corporate structure.
By analysing our data through this framework, we were able to show how the blurring and redrawing of legal boundaries through labour hire has clearly affected the workplace as a scale of justice for trade unions in the Norwegian transport and logistics sector. In part, this has to do with the variegated forms of externalisation currently practiced in the sector. Long-term and unpredictable labour hire practices place externalised workers in transitional positions where they have to negotiate their outsider status and prove their employability for the firm and their colleagues. In the Oslo region, a networked recruitment system based on Swedish labour migrants, mediated via temporary work agencies, encourage workers to work their way through levels of employment insecurity in order to secure permanent employment – a system that seems to enjoy a level of consent even among permanent workers and trade union representatives.
Still, there is room for manoeuvre, and this article has shown how individual, habitual and collective expressions of worker agency were articulated in the workplace. Collegial relations infused the labour process with peer pressure, competition and camaraderie negotiating the firmness of legal boundaries created by externalised employment. Moreover, workplace trade unions represented a collective dimension of agency, which in some cases renegotiated the boundaries created by management’s employment practices and in other cases cemented divisions between a core of permanent employees and a flexible periphery of workers. By comparing the four warehouses in this study, it is clear that labour hire is not simply introduced by management in a frictionless manner. Rather, it develops in interaction with local workplace cultures, which to different degrees accept or resist the construction of legal insiders and outsiders.
In three of the four warehouses, the local trade unions were unable to resist the extensive use of TAWs effectively. This is perhaps indicative of a wider problem for the Norwegian trade union movement. During a time when the labour market is undergoing rapid changes, they have relied extensively on their capacities to challenge the deregulation of labour law national scale and through resisting the implementation of EU legislation. As the outcome of regulatory change is manifested at the workplace scale, the trade unions need to develop their strategic capacities and levels of mobilisation in each warehouse in order to effectively resist flexible employment practices. Alternatively, the trade unions risk being reduced to co-managers of flexibilisation.
Trying to organise these workers via their direct employers, the temporary work agencies does not appear as a viable alternative. TAWs in Oslo warehouses very clearly express a lack of attachment and loyalty to their direct employer indicates that the staffing industry does not offer a sound basis for organising union members. In the eyes of the workforce, temporary work agencies are ‘place-less employers’ lacking the socio-material qualities that can foster a sense of community. Hence, the organisation of hired staff ought to go via the place of work.
Footnotes
Acknowledgements
The authors wish to thank editor Jamie Peck, two anonymous referees and Neil Coe for their constructive comments and suggestions on earlier versions of this text.
Declaration of conflicting interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The Welfare, Working Life and Migration (VAM) programme of the Research Council of Norway for the project “The Impact of Temporary Work Agencies on the Politics of Work” (Grant Number 227021), which this article is based on.